Prevention is better than cure: The role of infection prevention in the control of antimicrobial resistance

On the 18th of November, Europe will again be holding Antibiotic Awareness Day (EAAD). Antimicrobial resistance (AMR) is a topic high on national and international agenda at the present time and is likely to remain so for the foreseeable future. Resistance is rising, associated costs, both human and financial, are significant and there are no new antimicrobial classes riding over the horizon to reverse this situation. The question of ‘where does infection prevention fit into the AMR programme’ has never been more relevant. An antibiotic after all could really be considered to be a fire extinguisher that puts out the flames of infection; however, there are a number of criteria that have to be met before ignition takes place, many of which may be modified, or at least mitigated, in order to prevent infection and therefore the use of antibiotics. Colonisation by potential pathogens is a natural human state and infection could be considered (like a weed) to be an organism out of place. Prevention interventions are therefore twofold: to reduce the risk of transfer of pathogens to others and to protect the person from themselves.

When considering the routes of transmission from person to person, a number of interventions are possible, from advanced source control (Apisarnthanarak et al., 2014), through environmental hygiene (Anderson et al., 2017) to hand hygiene (Sickbert-Bennett et al., 2017). There are interventions at each point in the transmission journey, yet the evidence is that these are not reliably implemented (Anderson et al., 2011; Mitchell et al., 2015) leading to continuing risks to patients. Hand hygiene, despite healthcare providers reporting extremely high levels of compliance, is an area that still needs to be given attention as many practitioners are well aware of suboptimal compliance in their organisations.

Interventions aimed at protecting the patients from themselves has probably the greater evidence base, since there are less links in the chain to cause confounding. We have the evidence for interventions that demonstrate that the right practices, implemented consistently, for every patient will reduce infections and yet study after study demonstrates that even in research studies, interventions are not reliably implemented. A 2016 meta-analysis that examined the effectiveness of bundles has concluded that insertion and maintenance bundles for central venous lines are effective (Ista et al., 2016); however, these authors also decided to analyse reported compliance with the bundles of the included studies. They found that, even in research studies of bundle implementation, compliance was rarely reported, with over two-thirds of papers not reporting compliance at all, and of the studies that did report, it was suboptimal in every single one. Many of the studies included in the meta-analysis showed no benefit from the use of a bundle; however, in failing to report the compliance with the bundle, the validity of findings from studies that do not present compliance data could be questioned

Compliance with infection control first appeared in 1966 (Kunin and McCormack, 1966) and since that time the evidence base for effective interventions has steadily grown. We now know what to do (or what should be done); however, it seems that we do not yet know how to get fellow healthcare professionals to implement interventions of benefit to patients in a sustainable manner. The reasons for this require urgent study and those designing intervention studies should consider including implementation scientists in their research teams as the evidence to date is that we have either not been doing this well or that we are not reporting this aspect. The study previously mentioned (Ista et al., 2016) noted the high proportion of studies that have studied clinical outcomes of bundle interventions but not whether the interventions were actually implemented. Reviewers of papers submitted to journals should be asked to consider this aspect when considering if a paper is suitable for publication.

Studies of protocol compliance could precede further investigation of the reasons for non-implementation. Perhaps we have been using the wrong terminology. ‘Compliance’ is a passive, almost negative term, intimating that the person who is ‘compliant’ is acceding to the wishes and will of others. One definition of compliance is ‘a tendency to yield readily to others, especially in a weak and subservient way’ (http://www.dictionary.com/browse/compliance [accessed 3 October 2017]). In 1966, ‘compliance’ may have seemed to have been an acceptable way of describing whether staff followed a guideline or instruction. However, attitudes to being instructed on best practice in an era of increasing autonomy means that many senior healthcare professionals are unlikely to find this an attractive proposition. Perhaps use of an expression such as bundle observance would help improve perceptions, although involving clinicians in the process of bundle creation may also be beneficial, especially in choosing components where there are a large number of potential elements.

One study that attempted to use the Department of Health nine-point bundle (Department of Health, 2011) for surgical site infection (SSI) prevention (pre/intra/post phases of surgery) showed no impact of the bundle on SSIs, and when compliance was examined, less than one in five patients had received the whole bundle of care (Tanner et al., 2016). This may be due to insufficient engagement with staff around the development and implementation of the bundle, training issues or the availability of equipment when required. Studies that have included multidisciplinary stakeholder groups, steering committees, regular focus groups and newsletter updates have shown higher levels of observance of protocols (Crolla et al., 2012) and have shown a reduction in SSI that correlate with increases in adherence.

AMR is with us now, and the most effective way of reducing the risk of increasing morbidity and mortality caused by untreatable infections is to cut them off at source. We must concentrate efforts on convincing colleagues that interventions are effective. Perhaps by providing timely, focused, non-punitive and customisable data that may promote those at the clinical front line to question practices when outcomes are not good. Real and sustainable change rarely comes from a top–down approach and involving those who are to implement interventions is critical. After all, prevention is always better than cure; especially when there isn’t one.